github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9020] Llama3.2-vision Doesn't Accept an Array with Single Image. #5867

New Issue
Closed
opened 2026-04-12 17:12:25 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 17:12:25 -05:00
Owner
Copy Link

Originally created by @chigkim on GitHub (Feb 11, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9020

What is the issue?

If you pass an array containing single image to llama3.2-vision, it throws an error:
If you run the same request to minicpm-v, it works.
However, if you ask llama3.2-vision with the same requests but using OpenAI API, it accepts fine.
I understand that llama3.2-vision can handle only one image, but Ollama should be smart and accept array containing a single image.
Thanks!

Relevant log output

.venv\Lib\site-packages\ollama\_client.py", line 168, in inner
    raise ResponseError(e.response.text, e.response.status_code) from None
ollama._types.ResponseError: vision model only supports a single image per message (status code: 500)

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.5.7

Originally created by @chigkim on GitHub (Feb 11, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9020 ### What is the issue? If you pass an array containing single image to llama3.2-vision, it throws an error: If you run the same request to minicpm-v, it works. However, if you ask llama3.2-vision with the same requests but using OpenAI API, it accepts fine. I understand that llama3.2-vision can handle only one image, but Ollama should be smart and accept array containing a single image. Thanks! ### Relevant log output ```shell .venv\Lib\site-packages\ollama\_client.py", line 168, in inner raise ResponseError(e.response.text, e.response.status_code) from None ollama._types.ResponseError: vision model only supports a single image per message (status code: 500) ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.5.7
GiteaMirror added the bug label 2026-04-12 17:12:25 -05:00
GiteaMirror commented 2026-04-12 17:12:27 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 11, 2025):

llama3.2-vision does accept an array with one image via the ollama API:

 $ echo '{"model": "llama3.2-vision",
         "messages":[{
            "role":"user","content":"describe the animals shown in the images",
            "images": [
              "'"$(base64 puppy.jpg)"'"
            ]
          }],
         "stream":false}' | curl -s http://localhost:11434/api/chat -d @- | jq

{
  "model": "llama3.2-vision",
  "created_at": "2025-02-11T20:03:31.249468548Z",
  "message": {
    "role": "assistant",
    "content": "The image shows a small, white puppy sitting on what appears to be a stone or concrete surface. The puppy is positioned facing towards the right side of the image and has its ears folded back against its head. It is wearing a red collar adorned with a gold bell around its neck.\n\nThe puppy's fur is short and fluffy, giving it a soft and cuddly appearance. Its eyes are dark brown and appear to be looking off into the distance, as if it is gazing at something beyond the frame of the image. The puppy's nose is black, and its mouth is closed, giving it a peaceful expression.\n\nThe background of the image is out of focus, but it appears to be a room or outdoor space with a wall or other surface visible in the distance. The overall atmosphere of the image is one of calmness and serenity, with the puppy appearing relaxed and content as it sits on its perch."
  },
  "done_reason": "stop",
  "done": true,
  "total_duration": 10183230188,
  "load_duration": 2682934437,
  "prompt_eval_count": 19,
  "prompt_eval_duration": 2266000000,
  "eval_count": 185,
  "eval_duration": 5134000000
}
<!-- gh-comment-id:2651953204 --> @rick-github commented on GitHub (Feb 11, 2025): llama3.2-vision does accept an array with one image via the ollama API: ```console $ echo '{"model": "llama3.2-vision", "messages":[{ "role":"user","content":"describe the animals shown in the images", "images": [ "'"$(base64 puppy.jpg)"'" ] }], "stream":false}' | curl -s http://localhost:11434/api/chat -d @- | jq ``` ```json { "model": "llama3.2-vision", "created_at": "2025-02-11T20:03:31.249468548Z", "message": { "role": "assistant", "content": "The image shows a small, white puppy sitting on what appears to be a stone or concrete surface. The puppy is positioned facing towards the right side of the image and has its ears folded back against its head. It is wearing a red collar adorned with a gold bell around its neck.\n\nThe puppy's fur is short and fluffy, giving it a soft and cuddly appearance. Its eyes are dark brown and appear to be looking off into the distance, as if it is gazing at something beyond the frame of the image. The puppy's nose is black, and its mouth is closed, giving it a peaceful expression.\n\nThe background of the image is out of focus, but it appears to be a room or outdoor space with a wall or other surface visible in the distance. The overall atmosphere of the image is one of calmness and serenity, with the puppy appearing relaxed and content as it sits on its perch." }, "done_reason": "stop", "done": true, "total_duration": 10183230188, "load_duration": 2682934437, "prompt_eval_count": 19, "prompt_eval_duration": 2266000000, "eval_count": 185, "eval_duration": 5134000000 } ```
GiteaMirror commented 2026-04-12 17:12:27 -05:00
Author
Owner
Copy Link

@chigkim commented on GitHub (Feb 12, 2025):

Sorry about that! It looks like LlamaIndex duplicates image twice when using legacy ImageDocument for some reason.
Thanks for verifying!

<!-- gh-comment-id:2652334518 --> @chigkim commented on GitHub (Feb 12, 2025): Sorry about that! It looks like LlamaIndex duplicates image twice when using legacy ImageDocument for some reason. Thanks for verifying!
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#5867
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?